A growing number of common genetic variants have been robustly and reproducibly associated with type 2 diabetes (T2D). Despite these advances, the precise identity of the genes involved in increasing T2D risk has not yet been established. We propose to use pharmacogenetic and metabolomic approaches to inform the searches for causal variants and better define the molecular pathways involved. By challenging human subjects with a sulfonylurea or with glucose in the presence of metformin we hope to 1) distinguish responses depending on genotype at loci associated with T2D or related glycemic traits, or which impact metabolism of either drug; 2) confirm and expand an emerging metabolomic signature of insulin resistance by examining the human response to a glucose load with a larger panel of 380 metabolites (including lipid metabolites), 3) distinguish between the insulin and the glucose components of such response by discriminating the metabolomic profile in response to glucose versus the response to an insulin secretagogue; and 4) evaluate to what extent the refined metabolomic signature is correlated with genetic loci that predict insulin resistance. If successful, this proposal should help clarify the mechanisms by which genetic variants increase risk of T2D, and assess their impact on commonly used therapies. Whether the genetic defect is sufficient to prevent the expected action of either drug, or whether it can be overcome pharmacologically, would both be of clinical interest. In addition, this study should lay the groundwork for a longer outcomes-based pharmacogenetic trial.